Where Tuition Dollars Go, From Administrators to the Lazy River

Somewhere in America tonight, a seventeen-year-old is touring a campus and the guide is walking backward, as guides must, past a climbing wall, a juice bar and a river. Not a real river. A lazy river, a circular concrete channel in which a student may float in an inner tube while contemplating the fact that her parents are co-signing a loan. The guide does not mention the chemistry department. Nobody on the tour asks.

Folks, I want to talk about where the money goes. Not because I think a lazy river is a crime. A lazy river is a perfectly pleasant thing to float in. My complaint is with the billing. The family signs a check, or a promissory note, under the impression that it is buying an education, and what the bill actually funds is a much longer list. If you are going to pay for a river, you ought to know it is on the invoice.

Here is the plan. First, I define what we mean by tuition and by "instruction." Second, I walk through the three public data sets anybody can open to follow the money. Third, I do the arithmetic with a hypothetical university, labelled as a hypothetical, so you can see how a bond for a recreation center turns into a line on a student's bill. Fourth, I give the other side its best argument, because it is a better argument than the brochures make it sound. Fifth, I tell you what a family can ask before it signs.

Define the terms before you argue

Sticker price is the published tuition and fees. Net price is what a student pays after grants and scholarships. Instruction is, in the federal finance data, the spending categorised as instruction: faculty salaries and benefits, the cost of the classes themselves. It is only one line among many. The other lines include research, public service, academic support (libraries, deans' offices, computing), student services (admissions, advising, registrar), institutional support (the central administration, legal, human resources, fundraising), scholarships, and auxiliary enterprises such as dorms, dining and the recreation center.

That list is the whole argument in miniature. A family that says "tuition pays the professors" is describing one row of a table with a dozen rows. It is not a crazy assumption. It is what the word "tuition" suggests. But it is an assumption, and the data lets us test it.

Three public sources that let you follow the money

You do not need a leaked memo. You need three public things.

One: the Integrated Postsecondary Education Data System, known as IPEDS. It is run by the National Center for Education Statistics, the statistical arm of the Department of Education. Every college that takes federal student aid must report to it. Its Finance survey sorts a school's spending into the categories I just listed. Its Human Resources survey counts staff by occupational category, and the categories matter: instructional staff (faculty), executive, administrative and managerial staff, and a broad group of other professionals. Anyone can look up a given college and see how many people it employs in each category and how the spending is divided.

Two: the Delta Cost Project. This was a research effort housed at the American Institutes for Research that took the IPEDS files, cleaned them and published reports on college spending over several decades. Its reports are where many people first saw the shape of the story: instruction's share of spending flat or shrinking at many kinds of institutions, while the categories surrounding instruction, such as administration and student services, took a larger piece. I would tell you to read the reports themselves rather than my summary of them, and to notice that the authors were careful about how much a spending category can prove.

Three: the bond filings. Colleges borrow, and a college that borrows through the municipal bond market publishes documents describing what it borrowed and why. A new recreation center, a residence hall, a stadium renovation: each of these has a financing plan, and the financing plan has a payment schedule. That payment is called debt service, and it shows up in the budget whether or not a single student ever learns anything in the building.

Those are the three sources. IPEDS tells you who is on the payroll and where the money was spent. The Delta Cost Project tells you how the picture has moved over time. The bond filings tell you what the buildings cost to carry.

The amenities race, and the lazy river as its mascot

Now, the river. LSU's lazy river, shaped like the school's initials and attached to a large student recreation complex, became the emblem of a trend in press coverage that began in 2015 and continued after it opened in 2017, and the lazy river has been carrying the weight of the whole argument ever since. It is a little unfair to LSU, which is hardly the only school to have built one. The point is not that one campus indulged. The point is the logic of the race.

A college competes for students. Students, or more exactly their parents, tour campuses. A tour is a sales call, and on a sales call a climbing wall photographs better than a statistics classroom. So one school builds a rec center, the next school sees its admissions numbers and builds a bigger one, and the first school adds a river. Each school can say, truthfully, that it had no choice. Each school is right. That is how an arms race works, and it is why the people inside it so rarely feel like the villains.

Here is what I would ask a conservative to notice. This is a market. Nobody forced a student to choose the campus with the river. But it is a distorted market, because the buyer does not pay the full price at the time of purchase. A seventeen-year-old signs for a loan she will repay over a decade, subsidised and guaranteed by the federal government, and she is comparing campuses by what they look like and not what they cost over thirty years. When the buyer is not the payer and the payer is not the one choosing, you get more river than the buyer would choose if the money were hers. I would argue that is the single best explanation for the amenities race, and it is a thoroughly conservative explanation: it is about incentives, and about who bears the cost of a decision.

The arithmetic, with a hypothetical university

Let me do the math that nobody puts in the brochure. Everything in this section is a hypothetical. The numbers are round and invented for illustration, so that you can run the same arithmetic on a real college's real figures from IPEDS and its bond documents.

Lakeside State, a hypothetical. It enrols 10,000 students. It decides to build a recreation center and borrows $100 million, with a thirty-year bond at 5 percent interest, paid in equal annual installments.

The annual payment on a level-payment loan comes from a standard formula. At 5 percent over 30 years, each dollar borrowed costs about 6.5 cents a year to carry. So:

• $100,000,000 × 0.0651, roughly, equals about $6.5 million a year.
• Over thirty years, that is 30 × $6.5 million, about $195 million paid back on a $100 million building.
• Divide the annual payment by the students: $6,500,000 ÷ 10,000 = $650 per student per year.

Notice the second figure. Borrowing the money costs nearly as much again as the building. That is not a scandal, it is what interest is. But the family that took out a loan to pay the tuition is paying interest on the loan for the building and interest on its own loan, and neither is mentioned during the tour.

Now suppose Lakeside State's average net tuition, again hypothetically, is $10,000 a student per year. The $650 debt service is 6.5 percent of that $10,000. Not a fortune on its own. But the rec center is one project. A school in the race builds several: a dining hall, an upgraded dorm, a stadium suite, a student union. Four projects of this size mean $2,600 a student per year, a quarter of the hypothetical net tuition, before one professor has been paid.

The staffing arithmetic

Here is the second half of the picture, also a hypothetical. Suppose Lakeside State had 1,000 faculty and 1,000 non-faculty professional and administrative staff twenty years ago. Over twenty years, enrolment grows 20 percent, to 10,000 students. Faculty grows 10 percent, to 1,100. Non-faculty professional staff grows 60 percent, to 1,600.

• Students per faculty member: 10,000 ÷ 1,100 = about 9.1.
• Students per non-faculty professional: 10,000 ÷ 1,600 = 6.25.
• The ratio of non-faculty professionals to faculty went from 1.0 to 1,600 ÷ 1,100, about 1.45.

In this hypothetical the school employs nearly one and a half administrators and professionals per professor, where it once employed one. Is that bad? It depends on what the extra people do. I will come to the fair answer in a moment. But the pattern is the one the IPEDS Human Resources data lets any citizen test at a real school, and I would encourage you to do it, because the real numbers will not match my invented ones, and the difference between them is the story of your particular college.

Instruction as a share of the whole

Last, the number the whole post is built on. Suppose the hypothetical Lakeside State spends $400 million a year in total. If instruction is $120 million, then instruction is $120 ÷ $400 = 30 percent of the budget. A family writing a check that it believes is going to professors should be able to say: thirty cents of each dollar, in this example, reaches the classroom, and seventy cents goes to the rest of the table.

I want to be careful. That 30 percent is invented. The real figure differs by school, by sector, and by how the school books costs such as research and the medical center. It can be higher, and at some institutions it is. My claim is not that a particular number is true of all colleges. My claim is that the number exists, is public, and is almost never in the conversation when a family decides where to go. Look it up.

The best case for the other side

A fair argument deserves a fair opponent, so here is the best defense of the spending, as its defenders would put it.

One: administrators are partly a response to law. Colleges that take federal money must comply with a thick rulebook: Title IX offices, disability accommodation, financial-aid administration, reporting under federal statutes, and campus-safety reporting. Somebody has to do that work, and the somebody is staff. Much of the growth in non-faculty professionals is compliance staff, and compliance is something Washington requires, not something a college invented.

Two: students want services. Mental health counseling, tutoring, career services, advising. These are real services that real students use and that many parents would be upset to see cut. A school with no advising staff would be cheaper, and it would also lose students.

Three: state funding fell. At public universities, state appropriations per student dropped in many states, and tuition was raised to fill the gap. It is true, and I will not pretend it is not. That is a different story from the amenities story, and a family should know which story applies to its state.

Four: the amenities attract paying students. A rec center that raises enrolment can pay for itself. A school with an empty dorm is losing money. The defenders would say the river is a business decision, and sometimes a good one.

All four are real. I accept every one of them as true in some measure. What I do not accept is that they answer the question. They explain why spending grew. They do not explain why the family is not told, and they do not explain why the whole system keeps growing on the strength of borrowed money rather than on the strength of what students would voluntarily pay.

The answer, in four numbered points

1. Compliance explains some of the growth, and compliance is largely a federal creation. If conservatives want leaner campuses, the place to start is the rulebook that requires a staff member to administer each rule. A college can hardly be blamed for hiring the people the regulation obliges it to hire. Reform the regulation and the payroll follows.
2. The services argument cuts against the amenities argument, not for it. If the point of the extra staff is student success, then the families paying for it deserve outcome data: graduation rates, debt at graduation, repayment rates. Those figures are published by the federal government, and a school proud of its spending should be proud of them too.
3. Cheap, guaranteed credit feeds the race. When the federal government will lend nearly any amount to nearly any student, the college faces little price discipline. The more a campus raises its price, the more the loan system lets the student borrow. I would argue that is how a price spiral gets started and why it does not stop, and we take it up in the loan context in Grad PLUS Was a Blank Check for Graduate School, and Here Is the Proof.
4. Transparency is the minimum. A school ought to publish, on one page, the share of each tuition dollar that goes to instruction, and the debt service per student. If the number is good, the college has every reason to print it in bold.

What the Founders would make of it

This is a question about who pays and who decides, and the Founders cared about both. A republic runs on the principle that those who spend ought to answer to those who pay. The reason Americans in the 1770s objected to taxation without representation was not the size of the tax. It was that the spenders did not answer to the payers. A college system financed by borrowing that falls on a seventeen-year-old, guaranteed by a Treasury that falls on every taxpayer, and spent by administrators who answer to neither, has the structure of the thing the colonists disliked, though at a much lower temperature and with better food.

The same thread runs through the long history of public colleges in America. The land-grant colleges, which Lincoln, Justin Morrill and the Land Grant That Built the State University covers, were created to give working families practical learning at low cost, funded by land and used for farming, engineering and the mechanical arts. Nobody built them a river. The purpose was to teach, and the purpose was understood by the people paying.

I will also point you to the pattern in public schools, where The School Staffing Surge, Hiring Outpaced Enrollment for Decades documents a version of the same story: more hiring than enrolment growth called for, and a flat result. And for the broader question of whether spending buys results, Real Spending Per Pupil More Than Doubled Since 1970, So Where Are the Gains is the companion read. The same question applies to a university: when the dollars rise, what rises with them?

A word to the coverage

I have one more complaint, and it is about the way this story is told. When a school raises tuition, the headline says that "costs" rose, as though costs were weather. Costs do not rise. Costs are chosen. Someone decided to build the river, hire the vice president, and issue the bond, and each decision has a name on it. The coverage almost never asks who. It asks instead how families can "afford" the increase, which is the question the system most wants to hear, because it takes the increase as given and puts the burden on the family.

The better question is the one a conservative would ask of any government program: who decided, who pays, and who is accountable? Put those three questions to a university president and watch how quickly the lazy river turns into a policy discussion.

What a family can ask before it signs

You do not need an expert. You need five questions, asked of the admissions office and checked against the public data.

• What share of the operating budget goes to instruction, and has that share changed over the last decade? IPEDS has the Finance survey.
• How many full-time staff are in each category, and how has the number moved against enrolment? IPEDS has the Human Resources survey.
• How much debt does the college carry, and how much does debt service cost per student each year? The bond documents are public.
• What is the typical debt at graduation, and what share of borrowers are repaying? The federal College Scorecard publishes these.
• Which fees are mandatory, and what do they fund? The fee schedule is on the bursar's page and the answer is sometimes surprising.

None of this tells a family which college to choose. A college with a river and a strong engineering department might be the right choice. The point is that the choice should be made on the invoice, not on the tour.

The case in brief

1. Tuition is not a payment for teaching alone. Federal data sorts college spending into instruction and a dozen other categories, and instruction is only one row.
2. Public sources let any family check: IPEDS for staffing and spending, the Delta Cost Project's reports for the long trend, and bond filings for debt service.
3. The arithmetic is simple. In a labelled hypothetical, a $100 million bond at 5 percent over 30 years costs about $6.5 million a year, or $650 per student, and repays about $195 million in all.
4. The other side has real arguments: compliance burdens, student services and falling state funding. They explain the growth but not the silence about it.
5. Cheap, guaranteed credit lets colleges raise prices without losing buyers, which is why the answer is transparency and price discipline, not outrage.

What you can do: before you tour a college, pull up its entry in the federal College Scorecard, find its instruction share in IPEDS, and bring both numbers to the admissions office.